Carbohydrates: Edited by David D'Alessio and Luc TappyTight glucose control in intensive care units: an update with an emphasis on nutritional issuesElia, Marinosa,b; De Silva, AmindabAuthor Information aInstitute of Human Nutrition, University of Southampton, UK bSouthampton University Hospital NHS Trust, Southampton, UK Correspondence to Marinos Elia, Institute of Human Nutrition, University of Southampton, Southampton General Hospital, Mailpoint 113, Tremona Road, Southampton SO16 6YD, UK Tel: +44 2380 794277; e-mail: [email protected] Current Opinion in Clinical Nutrition and Metabolic Care: July 2008 - Volume 11 - Issue 4 - p 465-470 doi: 10.1097/MCO.0b013e3282fcea2a Buy Metrics Abstract Purpose of review Tight glucose control in ICU patients is now regarded as a goal of successful care. Some challenge this on the basis that it produces no benefit and may cause harm. We review the recent literature with an emphasis on nutritional aspects. Recent findings Since 2001, several randomized controlled trials have examined the effect of tight glucose control in ICU patients, but only one showed an overall survival benefit. Glucose potassium insulin infusions have also produced variable results, and sometimes cause falls in plasma phosphate with potential consequences. Several studies have shown tight glucose control is labour-intensive and increases the incidence of hypoglycaemia, which could have profound effects, especially if cerebral perfusion is poor. Nutritional intake during tight glucose control has generally been poorly defined. Unintentional cessation of nutrition has been identified as a risk factor for hypoglycaemia. No difference in glucose control has been found between parenteral and enteral feeding. Summary Without knowledge of nutrition provision in terms of carbohydrate, total energy intake and route of administration, some studies are difficult to interpret. It is currently difficult to recommend routine use of tight glucose control in the ICU. Many clinicians have adopted regimes to control glucose between 5.0–9.0 mmol/l. © 2008 Lippincott Williams & Wilkins, Inc.